Volume 85, Issue s1 - Supplement: XXXII. Messtechnisches Symposium 2018. Gastherausgeber: Andreas Fischer

In order to reduce the noise emission from airplane turbines perforated sheet with a cavity behind it (so called liner). The dampening principle of these elements are based on the interaction between sound and flow. Current physical models of this aeroacoustical sound absorption do not allow the development of more efficient liners, since the aeroacoustic interaction has not been yet fully understood. Therefore all relevant factors, like the local sound field distribution, have to be measured to fully comprehend the local sound absorption processes. For this purpose we present with this work a camera-based measuring system for the two dimensional non-invasive measurement of the local sound pressure with a high spatial resolution.

For a growing demand in reliability and precision for measurement results for surface metrology, multiple sensor systems or multisensor systems are increasingly used. By using sensor systems with differing characteristics, a more holistic result can be gained. In this contribution, two different methods for fusing data from confocal microscopy and white light interferometry are presented. The fusion itself is done by Gaussian processes and a weighted least squares method. For the Gaussian process, a hierarchical approch is used in combination with a Bayesian committee machine [20]. The weighted least squares method is realized with implicit surfaces and a kernel regression. In the following, the deployed sensors are shortly presented as well as the fusion methods, which are adapted for this use case. The results with simulated data as well as a real data set are discussed. Different metrics for the evaluation of the methods are presented.

Scanning white-light interferometers are widely used measuring instruments for 3D microtopography. The use of such devices in close-to-machine environments away from vibration-damped laboratories is due to environmental vibrations difficult or even impossible. In this article we present a passive vibration compensation method using the example of two interferometric sensors, which can compensate for the influence of arbitrary periodic and transient axial vibrations on the interferometric white-light measurements. Due to the temporally high-resolution distance detection of a distance measuring interferometer integrated into the white-light interferometer, all deviations from the desired linear depth-scanning process are measured and this information is used to correct the whitelight interference signals. From these, the surface topography is calculated by well-known evaluation algorithms. The vibration compensation is demonstrated based on comparative measurements with and without disturbing vibrations for different measuring objects.

Complex-shaped small-aperture optical components increasingly play a role in sensing and imaging applications. However, their metrological characterization remains challenging. A measurement method was developed, which bases on optical coherence tomography to measure form and centration of both functional surfaces of small-aperture lenses with only one measurement. Key challenge is the measurement of the functional surface facing away from the measurement head since it is distorted due to the refraction of light on the functional surface that faces towards the measurement head.

This article presents measurements of surfaces by performing closed-loop line scans using a nano coordinate measuring system (NMM-1) [4] and a new highfrequency focal-distance-modulated confocal point sensor. While maintaining the known advantages of the confocal measurement, the sensor represents an innovative combination of a fibre-coupled confocal illumination and detection with a tunable, acoustically driven gradient-index fluid lens (TAG-lens) [10] for modulation of the focus distance and a novel signal processing utilizing a lock-in amplifier. The new arrangement is able to achieve an approximately linear characteristic curve for the optimised feedback control of a CMS in scanning sample mode. This article emphasises the implementation and use of the sensor in a coordinate measuring system and the advantages of lateral closed-loop scans holding the distance between the sample and the objective constant compared to conventional lateral scans in axial stepwise layers.

Most common detectors in liquid chromatography use optical properties to detect analytes in the mobile phase. However, some analytes are not detectable with optical methods. Therefore, we introduce an easy to fabricate and easy to use electromagnetic detector principle which is based on a split-ring resonator. Very first measurements with different analytes confirm the basic functionality of the presented detector.

Hyperspectral images contain very useful information because of their high spectral resolution, which can be used for non-contact food testing. However, to process them with convolutional neural networks, large data sets are needed. This is especially true if the data is not preprocessed and therefore of high dimension. However, relatively few hyperspectral data sets exist. To solve this problem, the neural network can be pre-trained using an autoencoder, which compresses and reconstructs the image. By minimizing the reconstruction error, useful features can be learned to solve the original task. In this work, spice mixtures are used to investigate whether individual components can be detected. In particular, a neural network using a 3D convolutional autoencoder is trained with a small data set.

The method of intramolecular force compensation forms a new approach for reducing the response time of hydrogel-based chemical sensors. The principle is based on a bisensitive hydrogel, that fulfills both the function of a sensor and an actuator. This paper presents the measurement principle, the required bisensitive hydrogel and a sensor setup. The results show the desired bisensitivity of the sensor as well as the successfull reduction of the response time.

This article presents the special properties, operating mode, design, metrological properties, important application measurements of the interferometric atomic force microscope. The interferometric atomic force microscope serves as a probing system in the nanopositioning and nanomeasuring machine NMM-1. The main feature of the interferometric atomic force microscope is the position detector - the combined probe measuring system, which allows the simultaneous detection of torsion, bending and position of the cantilever by means of optical lever and interferometer. In this work the principle of simultaneous detection of vertical and angular position of the probe by means of an interferometer and an optical lever is explained, the optical arrangement is presented and afterwards the accomplished dimensioning of the position detector is discussed. This article deals also with metrological properties of interferometric atomic force microscope, calibration of its probing system and particularly with the uncertainty analysis of an example measurement.

The optical three-dimensional shape reconstrution of specular objects is still a difficult task. Deflectometric methods reconstruct the surface by observing the distorted image of a reflected reference scene. In this article a new approach is presented in which the deflectometric measurement method is carried out efficiently with a light-field camera. An interpretation of the camera as a massive multiple camera array allows to solve the deflectometric measurement problem with a multi-stereo approach. The performance of the surface reconstruction is confirmed in experimental measurements and it is shown that high accuracies can be achieved with the combination of deflectometry and light-field cameras.

Glaucoma prevention requires monitoring of the intraocular pressure, for which an acoustic self-tonometer has been developed. However, the laboratory tests on porcine eyes and the patient measurements in a clinical series of experiments prove significant cross sensitivity of the biometric parameters of the eye to the measured intraocular pressure. In order to consider the individuality of the eyes in the evaluation of the measured data, finite element simulations of the eye are carried out for different geometric characteristics. The influence of the eye geometry on the measurement uncertainty of the intraocular pressure to be measured is quantified on the basis of the simulation results. By knowing the individual eye geometry of the patient, the systematic measurement error of the intraocular pressure can be significantly reduced and a measurement uncertainty of the order of magnitude of current clinical tonometers can be achieved for the acoustic self-tonometer.

This article presents a novel type of damage detection system for power cables. The functionality of the measuring system can be verified with respect to macrobending-induced increase of attenuation. Due to the used integration approach, damage to the sheath at the power supply cable can be detected, whereby longterm failure causes can be detected at an early stage. By means of an integrated optical fiber and an optical time domain reflectometer, regardless of the electrical energy transmission in the cable, an exact and efficient damage localization can be provided, resuloting in time and costeffectiv procedure.

This manuscript describes a novel standingwave arrangement with two laser sources of different wavelengths, emitting towards each other. The resulting standing wave has a continuously moving intensity profile, a thin, transparent photo sensor is inserted into. When the sensor is moved along the optical axis a frequency shift, proportional to the velocity, occurs. This frequency shift can be evaluated for the purpose of interferometric length measurements. Zusammenfassung Dieser Artikel beschreibt einen neuartigen Stehende-Welle-Aufbau, bei dem zwei Laserstrahlen unterschiedlicher Wellenlänge in entgegengesetzter Richtung interferieren, wodurch eine heterodyne stehende Welle entsteht. In deren kontinuierlich bewegtes Intensitätsprofil wird ein dünner, transparenter Photosensor eingefügt. Bei einer Verschiebung des Sensors entlang der optischen Achse erfährt das Sensorsignal eine geschwindigkeitsproportionale Frequenzverschiebung, welche für interferometrische Längenmessung ausgewertet werden kann.

The definition and variation of input parameters of a model allows after a sample calculation the evaluation of his distributed output parameters. These are considered as reproducibility of this calculation. A thermal FEM model of a small multiple fixed-point cell for calibration of contact thermometers was simulated according to this procedure and validated with the reproducibility of measurements using a validation metric. After this a confidence interval in which predictions with the model at one not validated point could be carried out, was found. This approach is presented in this article.

Optical imaging is the building block of different measurement systems, whose imaging quality is influenced by many factors including the choice of imaging parameters, optics, and the imaging sensor. Realistic simulations can therefore be very beneficial for automatic optimization of such degrees of freedom. In this paper, we present a physically-based model for the image formation consisting of three parts: geometrical light transport simulation using available computer graphics methods, introducing lens imaging effects using Fourier optics, and integration of the EMVA 1288 sensor data. The proposed model has been utilized for simulating the images with speckle interference patterns caused by emitting a laser line on a metal surface. The agreement of the simulations with real images demonstrates that the model can produce accurate approximations to the real images, especially for diffraction-limited systems. We also discuss the simulation process for both coherent and incoherent imaging cases to simulate imaging systems having one or both types of illuminations.

Sound signals that correspond to discrete notes are separated from a polyphonic music signal including several instruments playing at the same time with a harmonic matching pursuit algorithm (HMPA). The required dictionary for the matching pursuit algorithm is developed in the first step of this method on the basis of the current music signal. In the second step, the note models that are stored in the dictionary are employed to approximate the music signal. The discrete notes can be extracted from this approximation and can be combined to separated melodies of occurrent instruments by means of previous knowledge about melodic lines.

Spectral information provided by hyperspectral images (HSI) enables to determine material abundances in mixtures. The acquisition and processing of HSIs is time-consuming and computationally intensive. For this reason, an approach based on spectral filtering is proposed. Instead of HSIs, the use of a spectral filters leads to the acquisition of a single intensity image. Depending on the design of the spectral filter, this intensity image holds information such as the abundances of a material. Within the scope of this work, we examine the design of spectral filters accounting for iterative approaches. Existing methods are investigated to be used for an optical determination of material abundances, and a new approach including several stages to estimate material abundances is presented.

In this paper an interferometric sensor is presented, which measures a mechanical strain with a differential laser Doppler measurement without contact on a technical surface and was designed for the fatigue strength test. The compact sensor detects the displacements at two adjacent measuring points in the plane perpendicular to the optical axis (in-plane), which is defined by a receiving lens. The differential displacement divided by the constant distance of two measuring points corresponds to the strain between them. The application of this strain sensor in the fatigue strength test requires a dynamic of elongation of 0.001 to 3 parts per thousand. A mathematical model was developed for the design of an optimal optical structure and for the determination of the theoretical physical resolution limit of the sensor. In this article we present our model and an optimized sensor design.

WirelessHART and WLAN and Bluetooth systems specially adapted to industrial conditions are examples of wireless communication systems that have become established for the networking of sensors and actuators in the industrial environment. In April of this year, an additional wireless standard IO-Link Wireless for discrete manufacturing automation has been introduced. Due to the global availability, which in turn leads to a high availability of low-cost RF communication components, the license-free ISM band in the 2.45 GHz range is frequently used extensively for communication. In order to avoid coexistence problems between the different systems, essentially listen before talk, frequency hopping and blacklisting methods have usually been employed. Spectrum analyzers have so far been used essentially offline as monitoring and diagnostic tools. This article proposes to integrate these online into the ongoing process in order to detect and predict possible interference between the communication systems competing for the radio channel. For this, they must be able to quickly and reliably detect and classify all known wireless systems. Furthermore, the price level of these spectrum analyzers must be considerably lower than that of laboratory measuring devices, so that they can be integrated in the sense of an additional sensor easily via known engineering tools in the communication.

The aerodynamic performance of wind energy rotor blades is negatively affected by erosion and contamination of the rotor blade leading edge. Conventionally used methods for boundary layer flow measurements at wind turbines are invasive and affect the flow itself. Therefore, a non-invasive method based on thermographic flow visualization is presented, which enables the quantification of the degree of influenced boudary layer flow due to erosion and contamination for the first time. The method can be used at real-scale wind turbines in operation. The localization of the laminar-turbulent transition and a geometric assignment of the two-dimensional camera measurement data to the three-dimensional rotor blade geometry are the basis for this purpose. An example measurement at a 1.5MW wind turbine reveals a contamination level of up to 5.2% with a measurement uncertainty of 0.1 %.